Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein the three-dimensional structure of the at least one real object is parameterized in form of a piece-wise planar structure.
2. The method according to claim 1 , wherein the parameter set is further including the three-dimensional rotation in a common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image.
3. The method according to claim 1 , wherein the parameter set is further including the vertical and horizontal focal lengths of the camera.
4. The method according to claim 1 , wherein the parameter set is further including the coordinates of the principal point of the camera.
5. The method according to claim 1 , wherein the density of the piece-wise planar structure is adaptively selected depending on the texture of regions of interests within the reference image.
6. The method according to claim 1 , wherein the pose of the camera when capturing the reference image is acquired using at least one of structured light, optical feature tracking, inertial sensor information or stereovision.
7. The method according to claim 1 , wherein the iterative minimization process comprises the step of: determining the update of the estimate of the parameter set until a certain number of iterations has been reached or the size of the updates falls below a defined threshold.
8. The method according to claim 1 , wherein for the determining of the depth of the at least first point of the at least one real object an update of the inverse of the depth is determined instead of an update of the depth.
9. The method according to claim 1 , wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization.
10. The method according to claim 9 , wherein the update of the depth or the inverse of the depth of the at least first point is only determined if the three-dimensional translation exceeds a certain threshold.
11. The method according to claim 9 , wherein the variance of the determined depth of the at least first point over time is determined.
12. The method according to claim 11 , wherein the force of the regularization on this point is increased depending on the variance.
13. The method according to claim 9 , wherein the update of the depth of the at least first point is regularized by using the distance to a reference value of this point which is computed as an average of the previously determined depth values of this point.
14. The method according to claim 9 , wherein the update of the depth of the at least first point is regularized by using the distance to a reference value of this point which is computed as a weighted average of the previously determined depth values of this point.
15. The method according to claim 14 , wherein the weights are defined by photometric or spatial information.
16. The method according to claim 9 , wherein the parameter set is including the depth of at least a second point of the at least one real object and in the iterative minimization process, the update of the depth of the first point and of the at least second point is determined until a certain number of iterations has been reached or the variance of the update for the depth of the first point or the variance of the update of the depth for the at least second point falls below a defined threshold.
17. The method according to claim 9 , wherein the parameter set is including the depth of at least a second point of the at least one real object and in the iterative minimization process, the update of the depth of the first point is determined until the variance of the update of the depth of the first point falls below a defined threshold, and the update of the depth of the at least second point is determined until the variance of the update of the at least second point falls below a defined threshold.
18. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein for the determining of the depth of the at least first point of the at least one real object an update of the inverse of the depth is determined instead of an update of the depth.
19. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization, wherein the update of the depth or the inverse of the depth of the at least first point is only determined if the three-dimensional translation exceeds a certain threshold.
20. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization, wherein the variance of the determined depth of the at least first point over time is determined.
21. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization, wherein the update of the depth of the at least first point is regularized by using the distance to a reference value of this point which is computed as an average or a weighted average of the previously determined depth values of this point.
22. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization, wherein the parameter set is including the depth of at least a second point of the at least one real object and in the iterative minimization process, the update of the depth of the first point and of the at least second point is determined until a certain number of iterations has been reached or the variance of the update for the depth of the first point or the variance of the update of the depth for the at least second point falls below a defined threshold.
23. A method for determining a parameter set which is designed to be used for at least one of determining the pose of a camera with regard to at least one real object or for determining a three-dimensional structure of the at least one real object, comprising the steps of: providing a reference image including at least a part of the at least one real object, capturing at least one current image including at least a part of the at least one real object, providing an initial estimate of a parameter set which is including at least the three-dimensional translation in the common coordinate system between the pose of the camera when capturing the reference image and the pose of the camera when capturing the current image, and the depth of at least a first point of the at least one real object in the common coordinate system, and determining an update of the estimate of the parameter set by means of an iterative minimization process, wherein in the iterative minimization process a first set of pixels in the reference image is compared with a computed set of pixels in the current image and the computed set of pixels in the current image used for the comparison varies at each iteration, wherein in the iterative minimization process the update of the depth or the inverse of the depth of the at least first point of the at least one real object is constrained by a regularization, wherein the parameter set is including the depth of at least a second point of the at least one real object and in the iterative minimization process, the update of the depth of the first point is determined until the variance of the update of the depth of the first point falls below a defined threshold, and the update of the depth of the at least second point is determined until the variance of the update of the at least second point falls below a defined threshold.
Unknown
December 8, 2015
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.